NWCHEM input file for CDFT on a Ethylene (C2H4) / C2F4 molecular pair for calculating charge transfer state energies + comparing to a TD-DFT method.

C2H4_C2F4_dist04.00.nw

title "C2F4 C2H4 with displacement: 04.00 Angstroms"

echo "This is a story about an ordinary teddy bear."

#Hellow 64GB / 16 node machine... will you be my friend?
memory stack 750 mb heap 1500 mb global 1500 mb

echo "Thundercats HO!"

geometry molA  units angstroms noautoz noautosym
#First our Ethylene... B3lyp/6-311g* optimisation, Gaussian, using symmetry. JMF 2013-01-28
C 0.000000 0.000000 0.663493
C 0.000000 0.000000 -0.663493
H 0.000000 0.922503 1.236446
H 0.000000 -0.922503 1.236446
H 0.000000 -0.922503 -1.236446
H 0.000000 0.922503 -1.236446
end

geometry molB
#Now our C2F2...  B3lyp/6-311g* optimisation, Gaussian, using symmetry. JMF 2013-01-28
C 04.00 0.000000 0.660905
C 04.00 0.000000 -0.660905
F 04.00 1.103866 1.386732
F 04.00 -1.103866 1.386732
F 04.00 -1.103866 -1.386732
F 04.00 1.103866 -1.386732
end

geometry molAB units angstroms noautoz noautosym
#First our Ethylene... B3lyp/6-311g* optimisation, Gaussian, using symmetry. JMF 2013-01-28
C 0.000000 0.000000 0.663493
C 0.000000 0.000000 -0.663493
H 0.000000 0.922503 1.236446
H 0.000000 -0.922503 1.236446
H 0.000000 -0.922503 -1.236446
H 0.000000 0.922503 -1.236446

#Now our C2F2...  B3lyp/6-311g* optimisation, Gaussian, using symmetry. JMF 2013-01-28
C 04.00 0.000000 0.660905
C 04.00 0.000000 -0.660905
F 04.00 1.103866 1.386732
F 04.00 -1.103866 1.386732
F 04.00 -1.103866 -1.386732
F 04.00 1.103866 -1.386732
end

basis
 * library 6-31++G*
end

#Fragments for molA
set geometry molA
dft
 xc b3lyp
 odft
 vectors input atomic output molA.mos
end
task dft ignore

#Fragments for molB
set geometry molB
dft
 xc b3lyp
 odft
 vectors input atomic output molB.mos
end
task dft ignore

#Ground state DFT please, for the dimer
set geometry molAB
dft
 mult 1
 vectors input fragment molA.mos molB.mos output molAB.mos #Fragment guess
 xc b3lyp
end
task dft energy

#Now do a TDDFT calculation based on this, solving for first singlet (~CT state)
tddft
 nroots 1
 singlet
end
task tddft energy

#OK; now let's try the big constrained calculation
set geometry molAB
dft
 vectors input fragment molA.mos molB.mos output molABcdft.mos #Fragment guess...

 iterations 50
 print  kinetic_energy

 xc b3lyp
 convergence nolevelshifting
#code for Constrained DFT - see http://www.nwchem-sw.org/index.php/Density_Functional_Theory_for_Molecules#CDFT_--_Constrained_DFT
 odft
 mult 1
# vectors swap beta 128 127  #unsure how to calculate this for PCBM... 231 alpha electrons      231 beta electrons <- mono PCBM
 cdft 1 6   charge +1.0 #Ethylene
 cdft 1 6   spin   +1.0

 cdft 7 12  charge -1.0 #C2F4
 cdft 7 12  spin   -1.0
end
set dft:cdft_maxiter 1000
task dft energy

non fragment guess version.nw

title "This is a story about an ordinary teddy bear."

echo "Thundercats HO!"

#Hellow 64GB / 16 node machine... will you be my friend?
memory stack 750 mb heap 1500 mb global 1500 mb

geometry units angstroms noautoz noautosym
#First our Ethylene... B3lyp/6-311g* optimisation, Gaussian, using symmetry. JMF 2013-01-28
C 0.000000 0.000000 0.663493
C 0.000000 0.000000 -0.663493
H 0.000000 0.922503 1.236446
H 0.000000 -0.922503 1.236446
H 0.000000 -0.922503 -1.236446
H 0.000000 0.922503 -1.236446

#Now our C2F2...  B3lyp/6-311g* optimisation, Gaussian, using symmetry. JMF 2013-01-28

C 10.000000 0.000000 0.660905
C 10.000000 0.000000 -0.660905
F 10.000000 1.103866 1.386732
F 10.000000 -1.103866 1.386732
F 10.000000 -1.103866 -1.386732
F 10.000000 1.103866 -1.386732

end
basis
 * library 6-31++G*
end

dft
 iterations 50
 print  kinetic_energy

 xc b3lyp
# convergence nolevelshifting
#code for Constrained DFT - see http://www.nwchem-sw.org/index.php/Density_Functional_Theory_for_Molecules#CDFT_--_Constrained_DFT
 odft
 mult 1
# vectors swap beta 128 127  #unsure how to calculate this for PCBM... 231 alpha electrons      231 beta electrons <- mono PCBM
 cdft 1 6   charge +1.0 #Ethylene
 cdft 1 6   spin   +1.0

 cdft 7 12  charge -1.0 #C2F4
 cdft 7 12  spin   -1.0
end
set dft:cdft_maxiter 1000

task dft energy

#tddft
# nroots 1
# singlet
#end
#
#task tddft energy

NB: I was stuck for ages due to not using a diffuse basis set, therefore the CDFT could not impose a suitable factor to cause the change in potential!
This NWCHEM script is probably overkill - you don't really need to do a fragment DFT setup for this calculation anymore, but I left it here as it might be a useful basis for doing charge transfer CDFT (diabatic states) calculations.

Thankyou Jarv from the past for your little NB - I've found myself doing exactly the same thing (too minimal a basis), and ended up Googling my way here.
The error you get is just the typical CDFT failed to optimise one: "CDFT failed to optimize multipliers"